Brian Kardell

If coding agents struggle to work with your code so will humans. Agent friendly code bases, are also human friendly. Human co-pilots use agents to make your code base more agent and human friendly.

44

2 Comments

OK, I am going to join the fray and weigh in about AI coding assistants. I have about a decade plus of enterprise software engineering experience under my belt, so I think I am competent to talk about this. THE GOOD: * Knowing a particular language or not no longer really seems to matter. I'm a bit rusty at Python. Instead of slowing me down, this literally seems to make zero difference when coding with Windsurf + some LLM (Gemini 2.5, Claude Sonnet, whatever) * Similarly, knowing the language's conventions don't seem to matter. "Make this more Pythonic" and ta da. Seriously, it's magical. * Test boilerplate is easy as pie. I really don't think there's much excuse not to write tests anymore. * Asking it to summarize what a really long class or something actually does is pretty nifty, especially in conjunction with reading through the code. * Commonly used things, like retries or exception handling or whatever, are absolutely effortless. * Rubber ducking with it is very nice. In the absence of co workers to talk to, it provides a really nice listening ear for brainstorming design ideas. It gives helpful feedback and suggests pre-existing libraries and similar open source implementations, saving buckets of time. * It really is autocomplete on steroids. Coding, especially with Windsurf, is an absolute breeze. THE BAD: * Abstraction. Hoo boy. It does not do well at this. Coding with any level of abstraction, especially when you're trying to abstract away hardcoded stuff and shift it to a config pattern is just.... ugh. It is like buddy programming with an incredibly enthusiastic but VERY dull-minded junior engineer. It's awful. If it does do abstractions, they're poorly chosen and clearly ripped off from someone else doing something barely similar. CONCLUSION: Well, that sounds like a lot of good and not very much bad, right? Yes, except the abstraction part is really important when it comes to building maintainable, modular, reusable code. And it's the hard part of coding, really. Does this revolutionize software development? Yes! But abstract thinking is fundamental to human intelligence. It's why our languages are full of metaphor. It's how humor works. And, unsurprisingly, it is how software development works.

520

72 Comments

Whenever you're writing something at work (whether it's a pull request, a document, a system design, a bug report, a feature request, etc.), make sure to give it a meaningful title and a TL;DR. Make it easy for folks to get a grasp of what it is you're doing.

52

5 Comments

This is quality work. So glad to see compiler team digging in and getting a richer picture than “build fast” or “build slow”. I definitely think we need to focus on the perception of rust as having slow builds (and the reality it’s based on), but to do that, we need to be clear on what workflows and use patterns we are targeting. The fix for fast CI turnaround is not the same as the fix for fast iteration on a broken test, and iterating on a test is not the same as debugging a type error. https://lnkd.in/eSKkNgXb

98